Volume 2013, Issue 1 408168
Research Article
Open Access

Commuting Quasihomogeneous Toeplitz Operator and Hankel Operator on Weighted Bergman Space

Jun Yang

Corresponding Author

Jun Yang

Department of Mathematics, Shanghai Maritime University, Shanghai 201306, China shmtu.edu.cn

Search for more papers by this author
First published: 28 July 2013
Citations: 1
Academic Editor: Giovanni P. Galdi

Abstract

We characterize the commuting Toeplitz operator and Hankel operator with quasihomogeneous symbols. Also, we use it to show the necessary and sufficient conditions for commuting Toeplitz operator and Hankel operator with ordinary functions.

1. Introduction

Let dA denote Lebesgue area measure on the unit disk š”», normalized so that the measure of š”» equals 1. For α > āˆ’1, we denote by dAα the measure dAα(z) = (α + 1)(1āˆ’|z|2) αdA(z). For 1 ≤ p < +āˆž, the space Lp(š”», dAα) is a Banach space. The weighted Bergman space is the closed subspace of analytic functions in the Hilbert space L2(š”», dAα). For each z ∈ š”», the application is continuous and can be represented as , where
()
This means that, if Pα is the orthogonal projection from L2(š”», dAα) onto , then Pα can be defined by
()
For a function f ∈ Lāˆž(š”», dAα(z)), we define the Toeplitz operator with symbol f by
()
It is well known that
()
Let U(α) : L2(š”», dAα(z)) → L2(š”», dAα(z)) be the unitary operator defined by , where f belongs to L2(š”», dAα(z)). Let g be in L2(š”», dAα(z)); we define a bounded linear operator Mg on L2(š”», dAα(z)) as follows:
()
Then we can define the small Hankel operator as follows:
()
as Hg = PαU(α)Mg.

The study of commuting Toeplitz operators on the Bergman and Hardy spaces over various domains and related operator algebras has a long lasting history. On the Hardy space of the unit disk, Brown and Halmos [1] first showed that two Toeplitz operators are commuting if and only if either both symbols of these operators are analytic, or both symbols of these operators are coanalytic or a nontrivial linear combination of the symbols of these operators is constant. On the Bergman space, the situation is more complicated. Axler and Čučković obtained the analogous result for Toeplitz operators with bounded harmonic symbols on the Bergman space of the unit disk [2]. The problem of characterizing commuting Toeplitz operators with arbitrary bounded symbols seems quite challenging and is not fully understood until now. In [3], Čučković and Rao used the Mellin transform to characterize all Toeplitz operators on which commute with for (m, p) ∈ N Ɨ N. Later in [4] Louhichi and Zakariasy gave a partial characterization of commuting Toeplitz operators on with quasihomogeneous symbols. Recently, Lu and Zhang [5, 6] characterized the commuting Toeplitz operators and Hankel operators with quasihomogeneous symbols. There are also many other important results [7–13]. Motivated by those works, we study commuting Toeplitz operator and Hankel operator on the weighted Bergman space. In this paper, we obtain the necessary and sufficient conditions for commuting Toeplitz operator and Hankel operator.

An operator that will arise in our study of Toeplitz operators is the Mellin transform, defined for any function φ ∈ L1([0,1], r dr); from the formula, is the Mellin transform as follows:
()
which is a bounded holomorphic function in the half plane {z : Re z > 2}.
Let φ ∈ L1(š”», dAα) be a radial function; that is, suppose that φ(z) = φ(|z|), z ∈ š”». In fact, if we define the function φr on [0,1] by φr(s) = φ(s), then a direct calculation shows that
()
so that if k ∈ ā„•,
()
Thus, Tφ is a diagonal operator on with coefficient sequence as follows:
()
This makes it relatively simple to work with the product of two operators with such radial symbols.
Now, we define the ā€œradializationā€ of a function f ∈ L1(š”», dAα) by the following:
()
It is clear that a function f is a radial if and only if rad(f) = f.
Let ā„œĪ± be the space of weighted square integrable radial functions on š”». By using that, trigonometric polynomials are dense in L2(š”», dAα) and that, for k1 ≠ k2, is orthogonal to , we see that is
()

Definition 1. Let φ be a function in L1(š”», dAα) which is of the form eikĪøf, where f is a radial function. Then one says that φ is a quasihomogeneous function of quasihomogeneous degree k.

A direct calculation gives the following lemmas which we will use often.

Lemma 2. Let p ∈ ā„• and let φ be an integrable radial function. Then,

()

Lemma 3. Let φ be an integrable radial function. Then, for p ∈ ℤ+,

()
and for p ∈ ā„•,
()

2. Commuting of Toeplitz Operator and Hankel Operator

Theorem 4. Let eipĪøf be a bounded function of quasihomogeneous degree p ≄ 0 and g = āˆ‘k∈Z eikĪøgk,α(r) ∈ Lāˆž(š”», dAα). Then if and only if the following conditions holds

  • (1)

    , if 0 ≤ k ≤ p āˆ’ 1 and j ≄ 0;

  • (2)

      , if k ≄ 0 and j ≄ 0.

Proof. For j ≄ 0,

()
If , then we have
()
which is equivalent to
()
that is,
()
From the aforementioned we get the following.

Case 1. For 0 ≤ k ≤ p āˆ’ 1 and j ≄ 0,

()
Case 2. For k ≄ 0 and j ≄ 0,
()
As a special case of Theorem 4, we can have the following corollary.

Corollary 5. Let f be a bounded radial function and g = āˆ‘k∈Z eikĪøgk,α(r) ∈ Lāˆž(š”», dAα). Then TfHg = HgTf if and only if

()
for k ≄ 0 and j ≄ 0.

Theorem 6. Let eāˆ’ipĪøf be a bounded function of quasihomogeneous degree āˆ’p < 0 and g = āˆ‘k∈Z eikĪøgk,α(r) ∈ Lāˆž(š”», dAα). Then if and only if the following conditions holds

  • (1)

    , if k ≄ 0 and p > j ≄ 0;

  • (2)

      , if k ≄ 0 and j ≄ p.

Proof. For j ≄ 0, we have

()

Then one has the following.

Case 1. For p > j ≄ 0,

()

Case 2. For j ≄ p,

()
If , then we have the following.

Case 1. For p > j ≄ 0,

()
Case 2. For j ≄ p,
()
That is one has the following.

Case 1. For p > j ≄ 0, k ≄ 0,

()
Case 2. For j ≄ p, k ≄ 0,
()

Theorem 7. Let eāˆ’isĪøg be a bounded function of quasihomogeneous degree āˆ’s ≤ 0 and f = āˆ‘k∈Z eikĪøfk,α(r) ∈ Lāˆž(š”», dAα). Then if and only if or

  • (1)

    , if s ≄ k ≄ 0 and j > s;

  • (2)

      j + 2), if s ≄ k ≄ 0 and s ≄ j ≄ 0;

  • (3)

    , if k > s and s ≄ j ≄ 0.

Proof. For j ≄ 0, we have the following.

Case 1. For j > s,

()
Case 2. For s ≄ j ≄ 0,
()
Applying , we have the following.

Case 1. For j > s,

()
Case 2. For s ≄ j ≄ 0,
()
If , then the equation holds.

Otherwise , we have the following.

Case 1. For j > s,

()
Case 2. For s ≄ j ≄ 0,
()
that is one has the following.

Case 1. For s ≄ k ≄ 0, j > s,

()
Case 2. For s ≄ j ≄ 0,
()
Then we get the following.

Case 1. For s ≄ k ≄ 0, j > s, .

Case 2. For s ≄ j ≄ 0,

()
Case 3. For k > s, s ≄ j ≄ 0,.

Corollary 8. Let g be a bounded radial function and f = āˆ‘k∈Z eikĪøfk,α(r) ∈ Lāˆž(š”», dAα). Then TfHg = HgTf if and only if

  • (1)

    ;

  • (2)

    and , for k > 0.

Finally, we will investigate the situation that both functions are ordinary functions.

Theorem 9. Let f = āˆ‘k∈Z eikĪøfk,α(r) ∈ Lāˆž(š”», dAα) and g = āˆ‘l∈Z eilĪøgl,α(r) ∈ Lāˆž(š”», dAα). Then TfHg = HgTf if and only if

()
for m ≄ 0 and n ≄ 0.

Proof. For j ≄ 0, we have

()
From the aforementioned we get, for m ≄ 0 and n ≄ 0,
()
If HgTf = TfHg, then we get
()
for m ≄ 0 and n ≄ 0.

The converse is easy to get.

Acknowledgments

The author would like to thank the referees for their valuable comments and suggestions, which helped to improve the paper. This work is supported by the National Natural Science Foundation of China (no. 11126061), Innovation Program of Shanghai Municipal Education Commission (no. 13YZ090) and the Science & Technology Program of Shanghai Maritime University (no. 20120098).

      The full text of this article hosted at iucr.org is unavailable due to technical difficulties.